Proposed Student Digital Twin Data Model and System Architecture for Personalized Learning

Education systems worldwide are tending to replace the standardized learning techniques with a more customized, student-tailored approach, in order to allow students advance according to their own capabilities and circumstances. Personalization of learning can be improved by integrating the concepts of human digital twin (DT), which is based on real-time data, together with artificial intelligence (AI). This integration enables the use of real-time data about students, which improves the understanding of their needs. Therefore, a more effective personalized learning experience that tends to be more adaptive could be achieved. This research proposes an architecture for a system that captures real-time data from students using data capturing tools and IoT devices, analyzes these data, and give real-time feedback to guide students during their learning process. The main objective is to decrease the percentage of student's failure rate and highlight students with excelling performance. In this research, the set of attributes that properly defines a digital real-time copy of a student are defined, the tools that can capture these attributes are determined, a model of a digital student in a learning environment is developed, and the architecture of the overall system, and its potential uses are discussed. Finally, an experiment is conducted to test the proposed student digital twin model, and system architecture and results are highlighted.